Executable Graph Loading With Contracts for Low-Latency Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing n-tier architectures separate data storage from processing logic, leading to impedance mismatches and reduced flexibility, extensibility, and responsiveness, which is detrimental in time-critical applications.
Innovation Solution
An executable graph-based model integrates data and processing logic within a single model, using contracts and scopes to manage access and maintain separability, allowing dynamic combination at run-time while preserving offline separability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If data and processing logic are separated into different tiers (n-tier architecture), then data management and processing become more organized and maintainable, but processing latency increases and system flexibility decreases
Solution Approach 1:
The patent combines data and processing logic into a unified graph-based model where nodes represent both data entities and their associated processing functions. This merging eliminates the need for separate data access layers and processing logic layers, allowing data and computations to coexist and interact within the same structure, thereby reducing processing latency while maintaining organizational benefits through the graph's inherent structure
2Ease of manufacture
If data and processing logic are separated into different tiers, then system maintainability improves, but system flexibility and extensibility deteriorate
Solution Approach 1:
The patent segments the system into reusable graph nodes and node types that can be independently defined, configured, and maintained. Each node represents a discrete unit of data or processing logic that can be modified, extended, or replaced without affecting the entire system. This segmentation enables both maintainability through modular design and flexibility through easy reconfiguration and extension of individual nodes
Solution Approach 2:
The graph-based model allows dynamic configuration and modification of the system structure at runtime. Nodes and their relationships can be added, removed, or modified without requiring system-wide changes or complex deployment processes. This dynamic nature enables the system to adapt to changing requirements while maintaining organizational structure through the graph's inherent hierarchy and relationships
3Ease of manufacture
If strict separation between data and processing logic is enforced, then architectural clarity improves, but responsiveness and efficiency of data processing decrease
Solution Approach 1:
The patent merges data and processing logic into a unified graph structure where nodes simultaneously represent data entities and their associated processing functions. This allows the system to maintain architectural clarity through the graph's organized structure while achieving high processing efficiency by eliminating the need for data to be transferred between separate layers and enabling direct in-graph computations
Data Source
AI summary
A method for contract-based loading of executable-graph based models is provided. A stimulus and an associated context are received. The stimulus is associated with a contract node having a contract identifier. Based on the contract identifier, a contract node manifest associated with the contract node, a storage location associated with a contract node state, and node identifiers of nodes associated with the contract node are obtained. The contract node is generated based on the contract node manifest and the contract node state. Node manifests associated with the nodes are obtained. Each node manifest comprises a node identifier and a storage location associated with a respective node state. Node states for the nodes are obtained. Each node state is obtained from the storage location associated with the respective node state. The nodes are generated based on the node manifests and the node states.